the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Jan 2025
Vertical and horizontal variability and representativeness of the water vapor isotope composition in the lower troposphere: insight from Ultralight Aircraft flights in southern France during summer 2021
Abstract. The isotopic composition of water vapor is a valuable tool to track atmospheric hydrological processes and to evaluate numerical models simulating the water cycle. To ensure accurate model-observation comparisons, understanding the spatial and temporal distribution of water vapor isotopes in the troposphere is crucial. The challenging task of obtaining highly resolved water vapor isotopic observations is typically addressed through airborne measurements performed onboard conventional aircrafts, but these offer limited microscale insights. This study utilizes observations from ultralight aircraft to examine the water vapor isotopic composition in the lower troposphere of southern France during late summer 2021. By combining the observations with conceptual and numerical models, we identify the main processes driving vertical and spatial variability of isotopic composition and we highlight the detection of short-lived, small-scale processes. The key findings of this study are that (i) at the hourly and sub-daily scales, vertical mixing is the dominant process affecting isotopic variability in the lowermost troposphere and boundary layer above the study site; (ii) evapotranspiration significantly impacts the water vapor isotopic signature, as revealed by the δ18O-δD relationship; (iii) measurable structures of the water isotopic fields emerge on the scale of 100s of m. The latter are particularly evident for δD, which also exhibit the largest differences in horizontal and vertical gradients. When combined with other airborne datasets, our results support a simple model forced with surface observations to simulate the vertical distribution of tropospheric δD, enhancing the comparison between surface observations and satellite data.
- Preprint
(7112 KB) - Metadata XML
-
Supplement
(4872 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2024-3394', Anonymous Referee #2, 28 Jan 2025
-
RC2: 'Comment on egusphere-2024-3394', Adriana Bailey, 08 Mar 2025
Vertical and horizontal variability and representativeness of the water vapor isotope composition in the lower troposphere: insight from Ultralight Aircraft flights in southern France during summer 2021
This study uses unique ultralight aircraft measurements in southern France to examine the horizontal and vertical variability of the stable water vapor isotope composition in the boundary layer and in the lowermost free troposphere. Evidence of coherent isotopic structures on the scale of 100s of meters are identified and discussed. The study also finds that vertical mixing is the dominant process that influences the boundary-layer water vapor isotope ratio on short timescales. A simple two-parameter model for estimating the isotopic profile of the boundary layer is presented based on these results.
The technical elements of this work are fascinating. And, the analysis is thorough and well structured, but the paper is also dense and long. I feel this work would reach a larger audience if the importance of identifying isotopic coherent structures could be conveyed more clearly. Indeed, a “key point” for each subsection would go a long way in helping bring the reader along with the manuscript.
So too, I think the paper requires some discussion on the generalizability of the findings to other atmospheric conditions and climatic environments. Figure 14 begins to address this question but only scratches at the surface: the flights considered represent a limited zonal band, and descriptions of the flight conditions (e.g. are they convective? quiescent?) are missing.
Additional, specific suggestions follow.
L 37 - To say that a process affects the stable isotopic ratio of water at the molecular level is a bit odd. The isotope ratio is necessarily a bulk feature of a large collection of molecules, it is not a property of a single molecule.
L 56 - “Specifically, the extent to which water vapor concentration and isotopic composition can resolve different atmospheric processes is still unclear.” Does this mean in terms of measurements or expectation? I would argue that theory largely guides us in what to expect and that the question of interest is whether we can measure it.
L 62 - I don’t agree that satellites generally provide breakthroughs on small scale and short lived processes. Their large footprint and infrequent sampling typically make them better suited for larger scale questions.
For Fig1, I suggest adding place labels for the key locations referenced later in text.
Section 2.4 and Caption of Figure 3 could be made more accessible and clear. What is the key message of this subsection? I assume the data are corrected for the time response, but this is not actually stated. Also, how does the time response change with water vapor concentration?
L340 - The paper states “blue and orange circles” but possibly blue and light green are meant instead?
L 366- “…Such a positive correlation might indicate the imprint of a local evapotranspiration signal in the boundary layer moisture.” It would be helpful to explain why this is the case, since there are two ways in which low slopes can be achieved. Either dD can decrease slowly relative to d18O, causing dD to fall above the MWL, or dD can fall below the MWL as the atmospheric moisture becomes isotopically heavier.
Section 3.4 could be simplified for broader accessibility. Line 391 is particularly confusing. Also, “correlation between standard deviation and vertical flight extent is high…” Does this simply mean that the aircraft measures a wider range of values as it traverses a larger vertical extent?
Section 3.5 would benefit from a statement of motivation and significance of results. Also, Figure 9, while technically interesting, is not particularly intuitive. What does it mean that Flight 12 has larger values than 5, for example? Why is this interesting given where these flights took place?
Fig 10 - I am confused about the axes of this figure. What are the values? Also, is the z scale the same as the x and y scales? It would be nice to get a sense of the size of the structures. Could the x and y axes be shown in units of m or km?
L 461 - “At L1200, close to the boundary layer height, the r2 significantly increases (0.83) and spatial features in the residual field are more evident (Fig. 10b).” I feel as though I’ve missed something here. If the r2 is high, meaning that the model predicts dD well, why are the residual features more substantial? It would help to provide a bit more physical context for what the results mean. Later, L 488 states: “the more evident the spatial features in the residual fields are, the smaller the r2.“ Are these contradictory statements?
L 494 - “Having seen that water vapor mixing ratio can provide a first-order approximation of the vertical and horizontal water vapor isotopic structure in the atmosphere.” To what extent does this depend on measuring in a fairly well-mixed boundary layer in the midlatitudes in September? Also, where are the model formulae stated?
Figure 12 - I worry that an RMSE of 20 permil in dD is about the same order of magnitude as the actual signal of isotope ratio change in the boundary layer (see Figure 6c). Also, d-excess is essentially constant in the boundary layer, and thus the prediction range is quite small. Knowing this, how meaningful are the simple models from whence the errors are derived?
L 560 - “Our results hence suggest that water vapor isotopes could be used as a proxy for studying boundary-layer development.” This assertion raises the question: why not just use water vapor, which can be measured at much higher temporal resolution? Would studying water vapor structures have given a different result?
L 582 - “a few data points within the boundary layer can be used to estimate the vertical profile of the water vapor isotopic composition”. But given the errors in Figure 12, could one simply assume a constant BL isotope ratio and obtain errors of equivalent magnitude? In that case, wouldn’t one data point suffice?
L 591 - The paper assumes a transpiration contribution. Couldn’t COSMOiso verify this? Evaluating such a hypothesis seems like an appropriate application for the comprehensive simulations.
Citation: https://doi.org/10.5194/egusphere-2024-3394-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 206 | 37 | 11 | 254 | 28 | 8 | 8 |
- HTML: 206
- PDF: 37
- XML: 11
- Total: 254
- Supplement: 28
- BibTeX: 8
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 87 | 36 |
| China | 2 | 29 | 12 |
| France | 3 | 27 | 11 |
| Germany | 4 | 17 | 7 |
| Switzerland | 5 | 13 | 5 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 87